Digitalización de activos críticos mediante IoT e inteligencia artificial para la predicción de fallas en una empresa de productos lácteos

Abstract

Abstract: This project presents the design and simulation of a predictive monitoring system for critical assets, specifically gear motors used in silo agitators, within the context of a dairy products company. The proposed system is framed within the principles of Industry 4.0, integrating Internet of Things (IoT) and Artificial Intelligence (AI) technologies to detect potential failures before they result in unplanned downtime. The system was developed entirely in a simulated environment using tools such as Node RED, InfluxDB Cloud, Grafana, and Python to emulate a complete flow of data acquisition, processing, storage, and analysis. By simulating variables such as temperature, vibration, and electric current, synthetic data was generated to train a supervised predictive model capable of classifying normal and fault conditions. The results validate the technical feasibility of the system, demonstrating that predictive maintenance solutions can be prototyped even in early stages or academic environments without the need for physical hardware. This approach provides a solid foundation for future real-world implementations.